Are We Obligated to Look for Trouble?眼前的落水儿童与孟加拉国的饥民——我们有找麻烦的义务吗？

By Timothy Huang

辛格认为对于眼前的落水儿童与远在孟加拉国的贫困人民应适用相同的原则，即“如果我们有能力避免坏事发生，而无需在道德上进行任何牺牲，我们就有道德义务去避免它”（1972:231）。他认为发达工业国家的人民有义务救助其他国家的穷人，正如他们有义务救助眼前的落水儿童一样（阿内森，2004）。这意味着，在某些情况下，道德要求我们就救助他人一事进行全球性的实质投入，而不局限于所在社区或政治国家（博尚，2016）。

辛格的比喻引起了较大争议。本文首先区分了眼前的落水儿童与孟加拉国的饥民之间就这一问题的根本性区别，在救助他们时，应当适用不同的道德准则。辛格误解了“牺牲”这一道德概念，为个体施加了过分的道德义务。救助义务是偶然产生的；没有任何人应该承担寻找其能救助的对象的义务。因此，辛格的理论缺乏说服力，这一过分的道德义务无助于帮助贫穷的人们。

落水儿童与孟加拉国济民之间的差异

辛格认为，人对于救助落水儿童和救助他国饥民负有同等义务，换言之，辛格似乎认为此二者之间的道德地位不存在差异。但是，这一论断是不合理的。基于以下原因，辛格未能对二者做出正确区分。

首先，辛格认为二者之间的唯一区别就是距离，这本身是值得怀疑的。他认为，“如果我们怀揣任何无差异、全球化和平等的理念，我们就不能仅仅因为一个人距离我们远而对其区别对待”（1974:232）。他假设，质疑其观点者仅仅是纠结于距离远近。但是，事实并非如此；而孟加拉国饥民的例子实际上相当复杂。例如，真正对孟加拉国人民的福祉负责的是孟加拉国政府，而非世界各地的道德的人民。如果当地人民挨饿，那么首先是当地政府未能履行职责。如果此时善良的世界各地人民站出来承担这一职责，那么当地政府只会更加不负责任。因此，帮助他们等于是在助长政府不负责任，尤其从辛格的功利主义视角来看，帮助他们是否具有道德上的可欲性，并没有清晰的答案。

另一个问题是道德成就感。如果某人救助了落水儿童，他知道他的极小牺牲拯救了孩子的生命；但是，如果他不亲自去孟加拉国，不亲自捐助有需要的人们，仅仅是把钱打给基金会，或者如同辛格的激进要求那般不再消费奢侈品，他永远不会知道他做了什么贡献。这种自己做出牺牲却不知道自己为谁做出了何种贡献的微弱的道德成就感显然与拯救落水儿童的案例无法比拟，也使得人们帮助孟加拉国饥民的正当性大大降低。

况且，有需求的人们距离遥远，本身确实是个大问题。例如，如果湿透了的衣服价值50英镑，那么此人便是付出了50英镑拯救了孩子的生命；但是，如果他捐助50英镑给孟加拉国的人民，这50英镑可能还不足以支付基金会运作这笔捐赠的必要成本。慈善组织本身也需要资金才得以运转。总之，辛格忽略了二者之间太多的区别。

第二，辛格混淆了两个相近却完全不同的概念：正确的事和有义务去做的事。毫无疑问，救助落水儿童和救助远方饥民都是道德上正确的事，但是人们并不必须承担做正确的事的义务——刺杀希特勒是正确的事，但我们不能因此事未成而苛责任何一位纳粹德国的公民。阿内森强调，道德正确并不意味着道德义务。一件事在道德上正确意味着支持它的道德理由胜过反对它的理由；但是做某件事的道德义务意味着未能完成时的道德惩罚（至少是愧疚和自责的情感）（阿内森，2004）。

基于此区别，做道德上错误的事并不必然导致这种愧疚或惩罚。一个人应当为没能向孟加拉国捐款而负责吗？显然不。一方面，辛格的理论基于结果主义的道德观，他必须考虑施以道德惩罚是否比其他选项结果更好；另一方面，因为远方的人未能尽到连当地政府和本国组织未能承担的义务而对其施以道德惩罚，本身便不能令人信服。相反，如果一个人在力所能及的情况下未对行将溺水的儿童施以援手，他必然为此受到道德的谴责：除他之外没有任何人有能力在现场能够救下孩子，惩罚他也是对其他应尽此道德义务者的警示。因此，拯救面前的落水儿童既是道德正确，也是道德义务，而拯救孟加拉国的饥民虽是道德正确，却非道德义务。

更何况，眼前的落水儿童是一个人分内之事——眼前的落水儿童迫使他做出选择：承担道德义务并拯救儿童，或是拒绝承担道德义务并放任其溺亡。站在河边看着孩子在河里挣扎的人必须在一瞬间作出决定，而这一决定也将成为其道德世界的一部分。换言之，当他发现了落水儿童，这就变成了他自己的事，尽管他有作出任何可能的选择的自由。相反，得知孟加拉国的饥民正在为基本的生存而斗争并不会影响一个人的道德世界——这一认识或许会令他感动，促使他做慈善，甚至改变他的世界观，但是他始终不需要就“做或不做”进行选择。远方的有需求的人们并不是我们道德世界的一部分。一个人固然可以选择去帮助他们，或者选择拒绝帮助他们；但是他也可以当做什么事也没有发生，而不必放弃任何道德义务——孟加拉国的贫穷与我们无关。

这一区分十分重要，因为它揭示了拯救眼前的落水儿童和帮助孟加拉国的饥民在道德上是完全不同的事。正如汉普顿的研究所揭示的那样，道德义务并不应该使人在压力下变成他人的仆人；“生活的艺术在于平衡各种道德义务——其中有些关乎你自己”（1993）。对于必须选择救或不救落水儿童的人来说，他们是在为自己抉择，而这一抉择本身会给他们带来道德满足；但是对于孟加拉国的饥民毫不关心并不会对自己的道德世界造成影响。因此，辛格脱胎于拯救落水儿童的道德义务无异于强制人们承担与之无关的事项。汉普顿问道：“为什么‘做道德的人’总是意味着对他人好——为什么它不能偶尔意味着对自己好？”（1993）这一问题指出了这个区别：在落水儿童案例中，做道德人是利己的，至少是道德满足、自我成就或是功利主义意义上的幸福；而在孟加拉国饥民案例中，做道德人则是纯粹利他的。

总而言之，辛格将放任孟加拉国饥民饿死比作放任眼前的落水儿童溺亡是不正确的，他忽视了这两个案例的不同的道德情形，导致整个比喻缺乏合理性。

过分的义务

辛格将孟加拉国的饥民比作面前的落水儿童是不合理的。这不仅仅是因为前文所述的二者的根本性差异，也是因为这一比喻给个人施加了过分的道德义务。

辛格的比喻和它蕴含的原则将会得出激进的结论：“每个人都有义务不购买奢侈品或改善性消费，并且自我克制，用积蓄帮助有需要的人”（米勒，2010）。例如，辛格反对购买“‘华丽’而非为了取暖”的衣服（1974:235）。这已经明显地以道德理由限制了处分自己财产的自由，甚至限制了政治自由和人权。即使以自己的积蓄进行捐助可能避免某些悲剧结果的发生，但是考虑一个人是否有义务做此事，却是一个完全不同的问题，因为这里不仅仅涉及效用（功利主义），也涉及财产权、道德自主性和分配正义。即使是单纯的讨论也不能对上述因素视而不见。正如米勒所言，辛格“在构建对人的平等尊重的要求时忽视了人与他人、自我、目标之间的关系在当中的角色”（米勒，2010:13）。对人的平等尊重并不意味着对仅知其贫穷的远方人民和眼前的落水儿童施以平等的道德关注；平等的尊重和相同的道德关注之间是有明确差异的。人人生而平等，无论距离远近，但辛格不能以此为基础得出结论，声称他们所有人都应当得到某个特定主体相同的道德关注。

辛格为个体设定的道德义务在直觉上是过分严厉的，超出了合理的限度，然而，限制此类为实现善而提出的要求仍然需要论证（卡甘，1984），而其中的某些论证并不成功（卡甘，1984；阿内森，2004）。下文将通过一个改版的比喻，证明辛格的比喻施加了过分的道德义务。

假设水性极好的某人住在滑铁卢车站附近，在威斯敏斯特工作。如果他在上下班路上遇到掉进泰晤士河的孩子，他就有救助他的道德义务。暂且不论孟加拉国的饥民，假设某天，他在油管直播上看到塔桥附近有儿童落水，他有救助这名儿童的道德义务吗？

如果他对此不承担道德义务，他就不可能对距离千百倍远的孟加拉国饥民承担救助的道德义务；如果他对此承担道德义务——他有义务拯救离家两英里远的落水儿童——那他又有什么理由不去拯救滨海绍森德的落水儿童呢？即使距离“不成问题”，辛格也不会支持此人跑如此远的距离去救孩子。原因很简单：附近有许多人比此人更有能力和机会救孩子，此人去救首先在在功利主义意义上是不合理的，这既减小了孩子的生存机会，也影响了施救者可以实现的成就。那么，问题在于，如果附近存在此道德义务负担者这一事实使得远距离施救成为功利主义意义上的坏的选择，那么，在当地政府、企业、NGO、以及乐于慈善事业的富豪面前，为什么辛格要求我们不远万里去帮助孟加拉国的饥民呢？辛格必须基于其功利主义视角和预设回答这一问题。

这揭示了辛格的牺牲原则至少从功利主义视角上是过分和不合理的，也说明他的比喻基于其自身的理论框架也难以令人信服。道德义务必须被限制在比这小得多的范畴内。

事实上，去帮助自己日常生活范围以外的人无异于沿着泰晤士河寻找是否有落水儿童可供施救——这很难与找麻烦区分开。拯救落水儿童是偶然的情境，我们不能忽视此类要求行为人牺牲的救助义务的偶然性。承担此类救助儿童的道德义务并不意味着承担寻找可能的落水儿童并施救的义务。一方面，“寻找”蕴含了主观的意愿，但是希望发现有人落水而需要救援，本身在道德上是错误的。这并不是说前述此人是主观上希望某个孩子落水；而是说，希望发现某人遭受厄运并成为自己承担和履行相应道德义务的情境是道德上错误的。否则，军工企业就应道德地希望本国陷入战争，从而承担抵抗侵略和打击敌军的道德义务，这是极为荒谬的——它们不过想发战争财而已。另一方面，救助义务既是不特定主体的道德义务，同时又是特定主体的法律义务。例如，河道行政管理部门对落水者负有法定的救助义务，落水孩子的监护人对救助他也负有法定的救助义务。换言之，道德义务只是偶然情境下的兼职工作，尤其在其他主体对此负有法定义务的情况下。当米勒提到，“从道德意义上讲，一个人能够保护小孩免受恶犬侵袭”（2010:12），他或许也忽视了孩子的父母和其他安保人员对此负有道德义务之上的、远超过“救助”的法定义务。因此，要求个体为此等事务牺牲日常生活以外的额外时间和精力是不合理的。

简而言之，作为道德义务的救助义务具有偶然性，我们没有找麻烦的道德义务。辛格关于面前的落水儿童和孟加拉国的饥民的比喻忽视了两个案例的重要差别，后者虽是道德上正确的，却不是道德义务。这一比喻导致了其对于道德牺牲的过分要求。他没有认真对待二者之间的差异，误解了道德关注的平等性。改版事例明确体现了辛格的比喻将使得其自身功利主义和结果主义的道德理论框架陷入矛盾。辛格关于拒绝奢侈品消费和不分情形的救助有需要的人的要求是过于激进的。

Peter Singer argues that the same principle should be applied between occurring a drown child nearby and the extremely poor people in Bangladesh, which is, “if it is in our power to prevent something very bad from happening, without thereby sacrificing anything morally significant, we ought, morally, to do it” (Singer, 1972: 231). He believes that the inhabitants of affluent industrialized societies bear the same moral obligation of assisting the poor in other societies as they face a drowning child nearby (Arneson, 2004), which implies that morality sometimes requires us to invest heavily in rescuing needy persons in the global population, not merely at the level of local communities and political states (Beauchamp, 2016).

Singer’s analogy proves to be quite controversial. This essay starts with the distinction between the drowning child nearby and the poverty in Bangladesh, and argues that these two cases should apply different moral principles, otherwise rescuing poor people in Bangladesh will have no plausible grounding in moral theory. Singer misinterprets his moral principle of sacrifice, and sets an over-demanding moral obligation for individuals. It follows that the rescue obligation only emerges on occasions; no one bears the obligation that consciously looking for the agents he can help, regardless of the sacrifice he might make. Accordingly, it concludes that Singer’s analogy between these two distinct cases is not convincing, and it is over-demanding to require individuals to sacrifice themselves in beneficence of the poor.

Distinction between the Drowning Child Nearby and the Poor People in Bangladesh

Singer believes that individuals bear the same obligation to rescue the poverty in Bangladesh as they save the drowning child. It seems to Singer that there is no difference in terms of moral status of people while facing the drowning child at hand and poverty around the world, but it is not plausible. Singer fails to distinguish these two cases in following senses.

    First, Singer’s belief that the only difference between these two cases is distance is quite problematic. He argues that “if we accept any principle of impartiality, universality, equality, or whatever, we cannot discriminate against someone just because he is far away from us” (Singer, 1974: 232), assuming those who question the same moral obligation to the poor in distance are considering the distance only. However, he is wrong about this assumption, and the Bangladesh case is quite complicated. For instance, it is the government of Bangladesh rather than the moral people around the world that bears the direct responsibility of the welfare of Bangladesh people. Now that people there are poor and dying, the local government fails to bear its own obligations to people at first. If foreigners come up and do this for the people, the government would bear less instead of more obligation in the future. Under this concern, that helping them may contribute to a bad government, it is not clear whether the aids are morally appealing, in particular, from the view point of utilitarianism that Singer supports. Another example is the feeling of moral achievement. If a man saves a drowning child, it will be clear that he sacrifices a small cost and saves this child; however, unless he goes to Bangladesh in person and give the money to the people in need, he will never know what exact contribution he has made by donating to a fund or, in Singer’s radical conclusion, not purchasing luxury goods. This moral relation, in which people sacrifice but do not know what exact contribution they have made, is undoubtedly different from the drowning child case, and provides much less moral justification to help the poor people in Bangladesh.

Besides, the mere fact that the people in need are “far from us” does matters, because it increases the cost of aiding. For example, if the wet clothes are £50, then the man costs £50 to save a child’s life; however, if he donates £50 to the people in distance, this £50 may not be enough for the necessary cost of operating this donation because the staff working for charity and the organisations are not themselves charity – they need fund to carry on. In general, Singer ignores too much about the complication of the Bangladesh case which makes it so different from the drowning child case.

Second, Singer mixes up two related but distinct conceptions: the right things and the things people are obligated to do. There is no doubt that saving a drowning child and help the poor people in Bangladesh are both morally right, but it is not necessary to follow that people are obligated to do these right things. Arneson argues that what is morally right is not morally obligatory. An act is morally right if the reasons for it overweigh the reasons against it; an act is morally obligatory just in case one should be punished (at the minimum, by guilt or self-blame) for failure to do it (Arneson, 2004). Under this distinction, it is clear that doing something morally wrong does not necessarily lead to this guilt. Should one be to blame if he does not donate to the poor in Bangladesh? Not necessarily. On the one hand, as Singer’s analogy is grounded in consequentialist morality, it is not surprising to ask whether punishing one would have consequences better than any alternative. On the other hand, it is not convincing to blame individuals in distance while the local government and native corporations in Bangladesh fail to bear their moral obligations in the first place.

On the contrary, one is definitely to be blamed if he does not save a drowning child before him whenever he is capable of this. Nobody else but the specific agent has the ability and capacity to save the child and punishing him in moral aspect for not doing so will be a warning for those who do not incline to bear this duty. Here is a coincidence and overlap of morally right and morally obligatory. That is, one is morally right and morally obligatory to save a drowning child, while he is morally right but not necessarily obligatory to help to poor people in Bangladesh. As a result, Singer’s analogy ignores the particularity of drowning child case, and fails to make it convincing.  

Moreover, a drowning child at hand is a factor of one’s own realm, which means he must make a choice: bearing this moral obligation and saving the child; or denying this obligation and letting the child die. The person standing by the water and watching the child drowning must make his mind in seconds, and his choice will become a part of his compliance of moral obligations. In other words, the moment he finds a drowning child, it has become his own thing, though in terms of pure moral obligation he is free to make any possible choice. On the contrary, knowing that there are people struggling for life in Bangladesh does not affect one’s self-regarding realm – the awareness of this fact may get him touched, promotes him to do some charity, even change the way he looks the world, but he does not need to make a choice of “do it or not”. That is, the people in need of aid but in distance is not part of the agents’ moral world. One can choose to help, or not help; but he can also do nothing without withdrawing any moral obligation he deserves – poverty in Bangladesh is none of his business.

This distinction is significant because it reveals that saving a drowning child at hand and helping the people in need in Bangladesh are totally different things. As Hampton observes, moral obligation is not something that makes individuals to become the servant of others under pressure; “[t]he art of living well is to know how to balance competing moral obligations – some of which are to yourself” (Hampton, 1993). For the agents who must choose to save the drowning child or not, they are making the decision for themselves, and the free choices make them morally satisfied; but not making a similar decision for the people in Bangladesh does not affect their own moral world, unless they are personally responsible for the poverty there. Thus, Singer’s analogy from the drowning child case is nothing different from forcing people to bear the moral obligations which are irrelevant to them. Hampton’s question that “why should ‘being moral’ always involve choosing in favour of other – why can't it sometimes involve choosing in favour of oneself?” (Hampton, 1993) points out this distinction: being moral in the drowning child case is in favour of oneself – at least, moral accomplishment, self-satisfaction or even utilitarian happiness; but being moral in Bangladesh case is purely in favour of others.

To sum up, Singer’s analogy between letting a drowning child die at hand and allowing others to die of poverty in Bangladesh ignores the different moral status of agents in these two cases, and accordingly fails to make this analogy plausible.

An Over-Demanding Obligation

It has been argued above that Singer’s analogy between a drowning child and poverty in Bangladesh fails to concern the significant distinctions between these two cases, which makes it implausible. His analogy is unconvincing not only because these two cases are too distinct, but also because it leads to an over-demanding principle of moral obligation.   

    Singer’s analogy and principle will lead to a radical conclusion: “everyone has a duty not to spend money on luxuries or frills, and to use the savings due to abstinence to help those in dire need” (Miller, 2010). For example, Singer disagrees with purchasing clothes “not to keep ourselves warm but to look ‘well‐ dressed’” (Singer, 1974: 235). This is obviously a limit on moral freedom of individuals, because it rules on the control of property for moral reasons. To some extent, it is even a limit on political liberty and human rights. Even if donating the saved money can prevent something very bad from happening,^[i] it is a different thing to consider whether the agents are obligated to donate it, because there is not only utility involved, but also property rights, moral autonomy and distributive justice. Even though a pure discussion in the realm of morality principle cannot take no account of these factors. As Miller observes, Singer “neglects the role of relationships to others, to oneself and to one's underlying goals in shaping the demands of equal respect for persons” (Miller, 2010: 13). With equal respect for all the human beings, it does not mean the equal moral concern to someone the agent knows nothing but poverty and a child drowning in his eyes – there are distance between equal respect and same moral concern. Just because all the humans are equal no matter how far they are from agents, Singer cannot conclude that all of them deserve the same moral concern from specific agents.

    While it seems intuitionally right that Singer sets too strict moral obligation for individuals which exceeds reasonable limits by this analogy, the reasons for limiting the requirement to pursue the good must be displayed (Kagan, 1984), and some of these attempts prove to be failure (Kagan, 1984; Arneson, 2004). Here follows a revised analogy which fills the logical gap of Singer’s analogy, and shows that Singer’s principle demands too much moral obligation.   

Suppose a person who lives in Waterloo Station and works in Westminster, so that if he meets a drowning child in Thames on the way to work, he is obligated to save the child. Put the poverty in Bangladesh aside, and imagine that one day he finds a drowning child near to Tower Bridge on YouTube while he is walking to work on Westminster Bridge, should he bear the obligation to save the child? It follows that, if he is not obligated to save the child outside his way, he will not be obligated to help the poor people in Bangladesh because they are further^[ii]; if he is obligated to save the child two miles away, why should not he save the drowning child in Southend-on-Sea? Granted that “far from us” does not matter, Singer will not support that a person going so far to save the drowning child – the reasons are obvious for him: there are persons nearby who are more capable to save the child, and requiring this person far away from the child is not utilitarian reasonable. Here comes the question: if there are obligation bearers nearby which makes the assistance in distance not a good utilitarian choice, why would Singer argue for helping the poor people in Bangladesh while there are local governments, corporations, NGOs and rich people who are in favour of charity? Singer must provide a satisfying answer to this question fully based on his assumption and utilitarianism.  

This reveals that Singer’s sacrifice principle is, at least, unreasonable in utilitarian views, and the demands which lead to the consequence that out of principle of utility is over-demanding in utilitarian views.^[iii] It also shows that Singer’s analogy is not convincing even in his own theoretical scheme, and the moral obligation must be limited to a much lesser extent.

In fact, assisting the people in need who are out of the agents’ normal life is nothing different from walking along with River Thames and looking for drowning children to save – it is hard to be distinguished with looking for trouble – rather than saving a drowning child on occasion. Bearing such rescue obligation does not mean the bearers are obligated to look for the objects they can rescue. On the one hand, “looking for” is with subjective expect, but it is morally wrong to expect some people to be in need of rescue. That is, it is morally wrong to expect something bad to happen so that there is context of bearing corresponding obligations. Otherwise, the munitions merchants are morally right to expect for unjust wars against their countries, so that they can bear their moral obligations of defending the nations, which is obviously ridiculous. On the other hand, rescue is a moral obligation for everyone, but it is also a legal obligation for specific ones. For example, there are particular departments of administration for the accidents and drownings in Thames. The point is that, moral obligation, especially where there are others bearing legal obligation on the same thing, is a “part-time job” and “on occasion stuff”. When Miller observes that “[o]ne can, in the morally relevant sense, prevent a child from being mauled by leading him away from a snarling dog” (Miller, 2010: 12), he might also ignore that only their parents and other guardians bear this professional and legal obligation beyond morality, and it is far more than “rescue”. As a result, it is implausible to require agents to sacrifice extra time or energy for those out of their daily life.

To sum up, Singer’s principle of sacrifice is over-demanding for both his opponents and himself as a utilitarian, and his analogy that leads to this demand is not convincing.

Conclusion

Singer’s analogy between letting a drowning child die in front of me and allowing others to die of poverty in Bangladesh ignores some significant distinctions between these two cases: the latter is much more complicated, unpredictable, and morally right but not necessarily obligatory, and it risks agents to be the servants of others. This analogy also results in his over-demanding set of the moral principle of sacrifice. He does not take enough consideration on moral distinction and mistakenly regard equality of human beings as justification of same moral concern from specific agents. The revised analogy above reveals that Singer’s analogy will result in paradox within his own moral basis of utilitarianism and consequentialism. Moreover, Singer’s analogy implies radical conclusion of not spending money on luxuries to save the money and assist the people in need and ridiculous conclusion of looking for drowning child around the world to save. Accordingly, Singer’s analogy is not convincing.

Bibliography

Arneson, R.J., 2004. Moral limits on the demands of beneficence? In The Ethics of Assistance: Morality and the Distant Needy. Cambridge University Press, pp. 33–58.

Beauchamp, T, 2016. "The Principle of Beneficence in Applied Ethics", The Stanford Encyclopedia of Philosophy (Winter 2016 Edition), Edward N. Zalta (ed.), URL = <https://plato.stanford.edu/archives/win2016/entries/principle-beneficence/>.

Hampton, J., 1993. Selflessness and the Loss of Self. Social Philosophy and Policy, 10(1), pp.135–165.

Kagan, S., 1984. Does Consequentialism Demand Too Much? Recent Work on the Limits of Obligation. Philosophy & Public Affairs, 13(3), pp.239–254.

Miller, R.W., 2010. Globalizing Justice: The Ethics of Poverty and Power, Oxford: Oxford University Press.

Singer, P., 1972. Famine, Affluence, and Morality. Philosophy & Public Affairs, 1(3), pp.229–243.

^[i] As argued in last section, the function of donating foreign poor people is still open to discuss.

^[ii] Note that Singer only considers the different distance as the distinction among the cases.

^[iii] It cannot be “insufficient-demanding” in this case.